dps8m-developers — Developers mailing list

Re: [dps8m-developers] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[Eric S. <er...@sw...>]

Thanks much, Gary. This is now installed on GHM and will be part of MR12.9.

Now when GHM starts up, IO.SysDaemon attaches rdra and GCOS.SysDaemon attaches rdrb. (Before, I had to change admin.ec’s read_cards entry to use rdrb since GCOS.SysDaemon could only use rdra).

> On Oct 1, 2025, at 12:03, gary_dixon@q.com wrote:
> 
> MCR10159 changes have passed audit and are ready to install.
> 
> Gary Dixon
> 
> 
>> On Oct 1, 2025, at 9:46 AM, Eric Swenson <er...@sw...> wrote:
>> 
>> I apologize if you receive this message more than once — I appear to be having an issue with my email sending.
>> 
>> This MCR is now approved. Gary, will you be willing to audit?  Audit directory is on GHM at >sysbuild>mcrs>MCR10159.  
>> 
>> I did update the MCR PDF to reflect the fact that the manifest constant SPLF should really be called SPNL, since NL is what Multics called what we now refer to as LF.
>> 
>> — Eric
>> 
>>> On Sep 8, 2025, at 3:52 PM, Eric Swenson <er...@sw...> wrote:
>>> 
>>> Hello Multicians:
>>> 
>>> Ticket https://multics-trac.swenson.org/ticket/381#ticket describes an issue with the “attach” command of the GCOS daemon.  This command doesn’t allow specification of the card reader to attach, and simply uses the “rdra” card reader.  Since this card reader may be in use by the IO SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon card reader driver is logged out.
>>> 
>>> This MCR proposes to address this problem by adding a parameter to the “attach” command.
>>> 
>>> Please review MCR10159 and comment on it or indicate your approval for this set of changes.  Provide comments or approvals by Reply-All to this email.
>>>  
>>> Thanks in advance for your review.
>>>  
>>> REVIEW DOCUMENT:         	MCR10159: https://s3.amazonaws.com/eswenson-multics/public/mcrs/MCR10159.pdf
>>> 
>>> REVIEW ENDS:             		Monday, August 22, 2025
>> 
> 

[dps8m-developers] Conversion of Multics Graphics Character Tables (GCT) to TrueType fonts (also, a MGS demo video)

[Jeff J. <tr...@po...>]

Greetings,

For another fun project, I wrote tools to convert the Multics
Graphics Character Table (GCT) format font files into modern
TrueType fonts, and documented the process in detail, which may
make for fun reading for those on this list:

https://gitlab.com/dps8m/font-gct-multics/-/blob/master/README.md

If want to see the completed TrueType fonts, there is a sample PDF:
https://gitlab.com/dps8m/font-gct-multics/-/raw/master/.preview/samples.pdf?inline=true

... or if you just want to download the fonts, grab them here:
https://gitlab.com/dps8m/font-gct-multics/-/tree/master/TrueType

I didn't actually check to see if the Multics versions of these
fonts have any differences from the "vanilla" Hershey vector font
sets, but, the point of the project was to convert Multics GCTs
to TrueType, and not to convert the Hershey vector fonts in
general.

If any Multics sites ever created custom fonts, these same tools
can be used to convert as well.  The tools in the repo should work
for any Mutlics GCT format font files, and not just the 12 fonts
that were included with the Multics Graphics System.  Alas, I fear
if any sites did create their own custom fonts, they are likely
long lost to time, but it's still worth mentioning.  Please get in
touch with me if you have other GCT fonts, even if just listings!

While a few tiny fixes will likely be applied soon (such as mapping
the "minus" glyph to the "dash" in the font tables), I believe the
conversion is of very high quality and extremely faithful to the
original fonts they would be drawn with a high quality plotter.

In related news, Eric Swenson recently recovered a demonstration
of the Multics Graphics System, which he plans to include in the
next Multics release.  I made a recording of this demonstration
using an emulated Tektronix 4015.  It features the fonts as well.

You can watch the video here - I recommend switching to HD 1440p:
https://www.youtube.com/watch?v=-Xyaz6Yk0wM

--
Jeffrey H. Johnson
tr...@po...

Re: [dps8m-developers] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[gary_dixon@q.com <gary_dixon@q.com>]

MCR10159 changes have passed audit and are ready to install.

Gary Dixon


> On Oct 1, 2025, at 9:46 AM, Eric Swenson <er...@sw...> wrote:
> 
> I apologize if you receive this message more than once — I appear to be having an issue with my email sending.
> 
> This MCR is now approved. Gary, will you be willing to audit?  Audit directory is on GHM at >sysbuild>mcrs>MCR10159.  
> 
> I did update the MCR PDF to reflect the fact that the manifest constant SPLF should really be called SPNL, since NL is what Multics called what we now refer to as LF.
> 
> — Eric
> 
>> On Sep 8, 2025, at 3:52 PM, Eric Swenson <er...@sw...> wrote:
>> 
>> Hello Multicians:
>> 
>> Ticket https://multics-trac.swenson.org/ticket/381#ticket describes an issue with the “attach” command of the GCOS daemon.  This command doesn’t allow specification of the card reader to attach, and simply uses the “rdra” card reader.  Since this card reader may be in use by the IO SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon card reader driver is logged out.
>> 
>> This MCR proposes to address this problem by adding a parameter to the “attach” command.
>> 
>> Please review MCR10159 and comment on it or indicate your approval for this set of changes.  Provide comments or approvals by Reply-All to this email.
>>  
>> Thanks in advance for your review.
>>  
>> REVIEW DOCUMENT:         	MCR10159: https://s3.amazonaws.com/eswenson-multics/public/mcrs/MCR10159.pdf
>> 
>> REVIEW ENDS:             		Monday, August 22, 2025
> 

Re: [dps8m-developers] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[Eric S. <er...@sw...>]

I apologize if you receive this message more than once — I appear to be having an issue with my email sending.

This MCR is now approved. Gary, will you be willing to audit?  Audit directory is on GHM at >sysbuild>mcrs>MCR10159.  

I did update the MCR PDF to reflect the fact that the manifest constant SPLF should really be called SPNL, since NL is what Multics called what we now refer to as LF.

— Eric

> On Sep 8, 2025, at 3:52 PM, Eric Swenson <er...@sw...> wrote:
> 
> Hello Multicians:
> 
> Ticket https://multics-trac.swenson.org/ticket/381#ticket describes an issue with the “attach” command of the GCOS daemon.  This command doesn’t allow specification of the card reader to attach, and simply uses the “rdra” card reader.  Since this card reader may be in use by the IO SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon card reader driver is logged out.
> 
> This MCR proposes to address this problem by adding a parameter to the “attach” command.
> 
> Please review MCR10159 and comment on it or indicate your approval for this set of changes.  Provide comments or approvals by Reply-All to this email.
>  
> Thanks in advance for your review.
>  
> REVIEW DOCUMENT:         	MCR10159: https://s3.amazonaws.com/eswenson-multics/public/mcrs/MCR10159.pdf
> 
> REVIEW ENDS:             		Monday, August 22, 2025

[dps8m-developers] Review of xcom.xpl Source for Multics

[gary_dixon@q.com <gary_dixon@q.com>]

Among the backup tapes from the MIT-Multics system, Eric Swenson found an early version of an XPL compiler for Multics.  According to the https://www.cs.toronto.edu/XPL/ web site, the XPL programming language is a derivative of the PL/I language designed for writing compilers in the late 1960’s (around the time PL/I was being designed).  XPL was first announced at the 1968 FJCC (Fall Joint Computer Conference) in San Francisco, CA.  The definitive description of XPL is the book  <http://www.amazon.com/exec/obidos/ASIN/0131550772/qid=962031185/103-2554179-0970214>A Compiler Generator <http://www.amazon.com/exec/obidos/ASIN/0131550772/qid%3D962031185/103-2554179-0970214> by William M. McKeeman, James J. Horning, and David B. Wortman (ISBN 13-155077-2) ("The Orange Peril") - see  <https://www.cs.toronto.edu/XPL/contents.html>Table of Contents <https://www.cs.toronto.edu/XPL/contents.html>.

In the 1960s and 1970s, MIT had a great interest in compilers for OS programming and computer applications.  Higher-level languages like FORTRAN, the Michigan Algorithmic Decoder (MAD), and PL/I itself were a few of the top choices at MIT in that time frame.

I got interested in writing simple compilers by an MIT course that introduced:
Productions as a means of describing valid phrases in a programming language; and 
Reductions as a means for assigning meaning to phrases written in that computer language.  

While working on Multics, I wrote the reductions translator which uses Reduction statements to translate statements of a source program into semantic actions that assign some meaning to each valid phrase found in the source.


This past history with translators spurred me to investigate the Multics version of XPL compiler (com.xpl) as devised in 1971-1973 time frame by:    Paul A. Karger, Fernando J. Corbató, and Jeffrey M. Broughton


I did a quick scan through xcom.xpl.  It is trying to translate XPL source directly into a non-reusable Multics-like object segment.  It is doing this using builtin-tables (created at Stanford, according to the commentary) mapping XPL language phrases/statements into some set of mid-level operations; then compiling representations of those operations into machine instructions on the target machine.

Although terms like H645 are used, I suspect it is generating instructions which are part of the pre-Extended Instruction Set (pre-EIS) instructions of the Multics 6090 or 6180.


The generated code is non-reusable in the sense that objects produced don't know how to interact with the Multics dynamic linker (at least full interaction is not attempted).  I suspect they operate more like GCOS object programs, except they are not encapsulated within a GCOS runtime environment which knows how to reload objects into a GCOS memory segment each time the program is referenced.  That's why they are non-reusable.  The generated objects appear to represent self-modifying text segments (at least to some extent).

I thought I might be interested in XPL becauses of its "compiler generator" capabilities.  But it appears to embody too much knowledge of its targeted 1970s runtime environment which would make true Multics operations difficult for an XPL-generated program.  It generates non-paged, non-segmented memory spaces containing mixture of code, space for XPL automatic variables, etc.

Instead of using memory-mapped access to segments, all segment reading and writing is done via I/O operations.  The xpl program is prepared to use multiple input and output files for differing kinds of data, and to enable dynamic loading of data into the limited memory space available to the generated object program, etc.  But all such extra files must be pre-attached by a hand-scripted pre-load_program before invoking xcom.  This preload program might be the xpl wrapper program which we hypothesize invokes the xcom object.  That seems like a valid hypothesis, based on my interpretation of xpl.info.


Given its limitations, I don't think I will spend much effort delving into the language.  I did a quick learning session on the BNF-like description of XPL on the Toronto U. web site.  XPL itself is a limited language with barely enough constructs to create compilers operating in a GCOS or IBM 360 runtime environment of the 1960s.

In a Unix or Multics environment, such programs can operate only with severe limitations on the OS-specific functionality they make available the compiler coder.  XPL provides a least-common-denominator set of facilities to the coder, to enable it to generate ostensibly portable programs that can be transferred between operating system environments.  But such transfers can only be done by hand-additions of OS-specific code_generator tables and information describing the target hardware architecture and operating system libraries and conventions.

In an IBM or GCOS era, its 1960 ideas were a good first start.  With addition of OS-specific services atop a vast variety of hardware architectures, the techniques embodied by XPL offer little help in creating transportable, robust, broad-service programs written in the XPL language.

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Jeff J. <tr...@po...>]

I assume that everyone who has searched for XPL has seen the
unofficial XPL homepage at https://www.cs.toronto.edu/XPL/. 

It has recently been updated.

There is a very cool XPL project that I've played with, which
is the Xerox Sigma XPL implementation, which has been ported
to UNIX systems:

https://sourceforge.net/p/xerox-xpl-compiler/code/ci/master/tree/

It requires the XPL to C translator to bootstrap, available from:

https://sourceforge.net/p/xpl-compiler/code/ci/master/tree/

The build does some tests that produce amusing benchmark output:

```
7230 cards containing 3038 statements were compiled.
[...]
Compilation rate = 43,380,000 cards per minute.
```

43.4 million cards per minute seems pretty impressive.

It would be an interesting project to see if the Xerox compiler could
be ported to Multics (or TOPS-20), bootstrapped from Multics XPL.

I haven't looked at the Multics compiler yet or the errors that you
are getting, but, for a user program, I can't imagine that it would
be coded in a way that it would explicitly depend on the 645 and not
work on the L68 or DPS8/M.

--
Jeffrey H. Johnson
tr...@po...

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[jim s. <jw...@jw...>]

Microdata had minicomputer system, the 3200 running  microcode 
implementing a machine the 32/s

They hired the UCLA group doing XPL to create a high level language 
using XPL for the system.  The resulting initial system was mostly 
written in XPL, first with cross compiling, then after the compiler 
(called MPL) was ported in MPL.

Genasys was the resulting system.  It was single user but allowed use of 
the system for customers.

They later embarked on a project to implement a system competitive in 
scope with other minicomputer multi user systems and called it Express.  
MPL was enhanced slightly and rebadged as EPL.

I worked on the OS part of that system as well as firmware to some 
degree, again slight mods from the 32/s.  We had language support, a 
pretty nice native file system with both random access as well as ISAM 
capabilities.

Was sold to Olivetti when Microdata ran into problems.

Thanks
Jim

More information here, see history document in this directory about Express.

http://www.bitsavers.org/pdf/microdata/express/


On 9/15/25 20:29, Daiyu Hurst wrote:
> XPL is kinda cool, it reminds me of Gary Kildal's PLM. I have a 
> version of XPL that translated XPL to C to build apps, and a MODCOM 
> III simulator written in XPL. Runs on Windows and Linux. I needed a 
> reference for the MODCOMP III to use as a guide to a possible 
> emulation of one that had been used as a front-end processor for the 
> CDC 6500 at Purdue. Their unique home-rolled OS came to us as source 
> code, and with some effort, I got it to build and run. It was the 
> first computer system I used, both punched-card and terminal input. 
> The terminal experience wasn't like that of Control Data's offerings.
>
> So you're saying this XPL is written in XPL? Maybe use this XPL-to-C 
> compiler to bootstrap it?
>
> -Dai
>
> On Mon, Sep 15, 2025 at 12:42 PM Eric Swenson <er...@sw...> wrote:
>
>     It’s going to take some work. Doesn’t compile. And I’m not sure
>     how to build the XPL compiler from source (xcom.xpl).
>
>>     On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>>
>>     Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.
>>      “Working on” means that I gather all the right bits, make sure
>>     it compiles and runs, updating it if necessary until it does. 
>>     Lots of the MIT-MULTICS code, based on MR10.2, needs updating for
>>     releases post-MR10.2.
>>
>>     I’ll reply to this thread regarding the results of working on XPL.
>>
>>     — Eric
>>
>>>     On Sep 15, 2025, at 9:25 AM, Dean Anderson
>>>     <de...@go...> wrote:
>>>
>>>     Eric:
>>>
>>>     I would be interested in the XPL related stuff. It was the first
>>>     "compiler generator" language I ran across and wanted to play
>>>     with badly but never had access to a system that ran it.
>>>
>>>     It's a stripped down version of PL/1 with some tools for
>>>     generating what are really translators rather than compilers
>>>     (though with the addition of semantic translation code you can
>>>     make a compiler with it).
>>>
>>>     Not great urgency but I'd love to play with it.
>>>
>>>     Thanks,
>>>     Dean
>>>
>>>     On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...>
>>>     wrote:
>>>
>>>         Attached you’ll find a list of all the programs installed in
>>>         the >aml directory on MIT-MULTICS at the time that that
>>>         system was shut down.  Those entries marked with INSTALLED
>>>         have been installed in the >aml for MR12.9 (not yet
>>>         released).  Those entries not so-marked have not been included.
>>>
>>>         I’m writing to provide anyone with the opportunity to lobby
>>>         for other programs to get integrated into the MR12.9 AML
>>>         library.  Please let me know if you see something here that
>>>         you really want installed in MR12.9 >aml.  Feel free to ask
>>>         me questions about these and I’ll endeavor to answer them
>>>         (undoubtedly it will require me to look at the sources/info
>>>         segments and attempt to rationalize not including them or
>>>         including them).
>>>
>>>         I will soon send out a similar list for the MIT-MULTICS SIPB
>>>         library, of which I’ve already installed a few programs into
>>>         the MR12.9 >aml.
>>>
>>>         — Eric
>>>
>>>         _______________________________________________
>>>         Dps8m-developers mailing list
>>>         Dps...@li...
>>>         https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>>
>>
>
>     _______________________________________________
>     Dps8m-developers mailing list
>     Dps...@li...
>     https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>
>
>
> _______________________________________________
> Dps8m-developers mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-developers

Re: [dps8m-developers] [Dps8m-users] MIT-MULTICS AML and MR12.9

[Tom P. <tom...@gm...>]

Aha. I worked on GCOS8 and but must have  forgotten that the code was
compiled on GCOS and not Multics.

As a student intern, I wrote a fair number of the PL6 macros and then wrote
some of the in-code formatted comments that were used to create the man
pages for the GCOS 8 system calls.

I worked for John Wertz, who was not a Multician, but "got around" or so I
was told.

But I was flipping from GCOS to CP6 to Multics several times a day, so
details have become fuzzy.

--tep




On Mon, Sep 15, 2025 at 9:36 PM Gary Dixon <gar...@ic...> wrote:

> CP-6 software was never part of a Multics Release, and did not run on
> Multics hardware.  It required use of the Climb instruction which was
> replaced by the Call6 instruction on Multics.  CP-6 had its own software
> release mechanisms, which did include the PL-6 compiler.  PL-6 was also
> used for some GCOS transaction processing software (IIRC).  So it might be
> available in GCOS software releases, or even on simulated GCOS systems
> running today.
>
> On Sep 15, 2025, at 9:22 PM, Tom Perrine <tom...@gm...> wrote:
>
> Did the PL6 compiler get saved somewhere? Is it on a regular Multics
> distro?
>
> --tep
>
>
> On Mon, Sep 15, 2025 at 6:30 PM Daiyu Hurst <dai...@gm...> wrote:
>
>> XPL is kinda cool, it reminds me of Gary Kildal's PLM. I have a version
>> of XPL that translated XPL to C to build apps, and a MODCOM III simulator
>> written in XPL. Runs on Windows and Linux. I needed a reference for the
>> MODCOMP III to use as a guide to a possible emulation of one that had been
>> used as a front-end processor for the CDC 6500 at Purdue. Their unique
>> home-rolled OS came to us as source code, and with some effort, I got it to
>> build and run. It was the first computer system I used, both punched-card
>> and terminal input. The terminal experience wasn't like that of Control
>> Data's offerings.
>>
>> So you're saying this XPL is written in XPL? Maybe use this XPL-to-C
>> compiler to bootstrap it?
>>
>> -Dai
>>
>> On Mon, Sep 15, 2025 at 12:42 PM Eric Swenson <er...@sw...> wrote:
>>
>>> It’s going to take some work. Doesn’t compile. And I’m not sure how to
>>> build the XPL compiler from source (xcom.xpl).
>>>
>>> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>>>
>>> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working
>>> on” means that I gather all the right bits, make sure it compiles and runs,
>>> updating it if necessary until it does.  Lots of the MIT-MULTICS code,
>>> based on MR10.2, needs updating for releases post-MR10.2.
>>>
>>> I’ll reply to this thread regarding the results of working on XPL.
>>>
>>> — Eric
>>>
>>> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...>
>>> wrote:
>>>
>>> Eric:
>>>
>>> I would be interested in the XPL related stuff. It was the first
>>> "compiler generator" language I ran across and wanted to play with badly
>>> but never had access to a system that ran it.
>>>
>>> It's a stripped down version of PL/1 with some tools for generating what
>>> are really translators rather than compilers (though with the addition of
>>> semantic translation code you can make a compiler with it).
>>>
>>> Not great urgency but I'd love to play with it.
>>>
>>> Thanks,
>>> Dean
>>>
>>> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...> wrote:
>>>
>>>> Attached you’ll find a list of all the programs installed in the >aml
>>>> directory on MIT-MULTICS at the time that that system was shut down.  Those
>>>> entries marked with INSTALLED have been installed in the >aml for MR12.9
>>>> (not yet released).  Those entries not so-marked have not been included.
>>>>
>>>> I’m writing to provide anyone with the opportunity to lobby for other
>>>> programs to get integrated into the MR12.9 AML library.  Please let me know
>>>> if you see something here that you really want installed in MR12.9 >aml.
>>>> Feel free to ask me questions about these and I’ll endeavor to answer them
>>>> (undoubtedly it will require me to look at the sources/info segments and
>>>> attempt to rationalize not including them or including them).
>>>>
>>>> I will soon send out a similar list for the MIT-MULTICS SIPB library,
>>>> of which I’ve already installed a few programs into the MR12.9 >aml.
>>>>
>>>> — Eric
>>>>
>>>> _______________________________________________
>>>> Dps8m-developers mailing list
>>>> Dps...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>>>
>>>
>>>
>>> _______________________________________________
>>> Dps8m-developers mailing list
>>> Dps...@li...
>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>>
>> _______________________________________________
>> Dps8m-users mailing list
>> Dps...@li...
>> https://lists.sourceforge.net/lists/listinfo/dps8m-users
>>
> _______________________________________________
> Dps8m-users mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-users
>
>
>

Re: [dps8m-developers] [Dps8m-users] MIT-MULTICS AML and MR12.9

[Gary D. <gar...@ic...>]

CP-6 software was never part of a Multics Release, and did not run on Multics hardware.  It required use of the Climb instruction which was replaced by the Call6 instruction on Multics.  CP-6 had its own software release mechanisms, which did include the PL-6 compiler.  PL-6 was also used for some GCOS transaction processing software (IIRC).  So it might be available in GCOS software releases, or even on simulated GCOS systems running today.

> On Sep 15, 2025, at 9:22 PM, Tom Perrine <tom...@gm...> wrote:
> 
> Did the PL6 compiler get saved somewhere? Is it on a regular Multics distro?
> 
> --tep
> 
> 
> On Mon, Sep 15, 2025 at 6:30 PM Daiyu Hurst <dai...@gm... <mailto:dai...@gm...>> wrote:
>> XPL is kinda cool, it reminds me of Gary Kildal's PLM. I have a version of XPL that translated XPL to C to build apps, and a MODCOM III simulator written in XPL. Runs on Windows and Linux. I needed a reference for the MODCOMP III to use as a guide to a possible emulation of one that had been used as a front-end processor for the CDC 6500 at Purdue. Their unique home-rolled OS came to us as source code, and with some effort, I got it to build and run. It was the first computer system I used, both punched-card and terminal input. The terminal experience wasn't like that of Control Data's offerings. 
>> 
>> So you're saying this XPL is written in XPL? Maybe use this XPL-to-C compiler to bootstrap it?
>> 
>> -Dai
>> 
>> On Mon, Sep 15, 2025 at 12:42 PM Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>>> It’s going to take some work. Doesn’t compile. And I’m not sure how to build the XPL compiler from source (xcom.xpl).
>>> 
>>>> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>>>> 
>>>> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on” means that I gather all the right bits, make sure it compiles and runs, updating it if necessary until it does.  Lots of the MIT-MULTICS code, based on MR10.2, needs updating for releases post-MR10.2.
>>>> 
>>>> I’ll reply to this thread regarding the results of working on XPL.
>>>> 
>>>> — Eric
>>>> 
>>>>> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go... <mailto:de...@go...>> wrote:
>>>>> 
>>>>> Eric:
>>>>> 
>>>>> I would be interested in the XPL related stuff. It was the first "compiler generator" language I ran across and wanted to play with badly but never had access to a system that ran it.
>>>>> 
>>>>> It's a stripped down version of PL/1 with some tools for generating what are really translators rather than compilers (though with the addition of semantic translation code you can make a compiler with it).
>>>>> 
>>>>> Not great urgency but I'd love to play with it.
>>>>> 
>>>>> Thanks,
>>>>> Dean
>>>>> 
>>>>> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>>>>>> Attached you’ll find a list of all the programs installed in the >aml directory on MIT-MULTICS at the time that that system was shut down.  Those entries marked with INSTALLED have been installed in the >aml for MR12.9 (not yet released).  Those entries not so-marked have not been included.
>>>>>> 
>>>>>> I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.  Please let me know if you see something here that you really want installed in MR12.9 >aml.  Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).
>>>>>> 
>>>>>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 >aml.
>>>>>> 
>>>>>> — Eric
>>>>>> 
>>>>>> _______________________________________________
>>>>>> Dps8m-developers mailing list
>>>>>> Dps...@li... <mailto:Dps...@li...>
>>>>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>>> 
>>> 
>>> _______________________________________________
>>> Dps8m-developers mailing list
>>> Dps...@li... <mailto:Dps...@li...>
>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>> _______________________________________________
>> Dps8m-users mailing list
>> Dps...@li... <mailto:Dps...@li...>
>> https://lists.sourceforge.net/lists/listinfo/dps8m-users
> _______________________________________________
> Dps8m-users mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-users

Re: [dps8m-developers] [Dps8m-users] MIT-MULTICS AML and MR12.9

[Tom P. <tom...@gm...>]

Did the PL6 compiler get saved somewhere? Is it on a regular Multics distro?

--tep


On Mon, Sep 15, 2025 at 6:30 PM Daiyu Hurst <dai...@gm...> wrote:

> XPL is kinda cool, it reminds me of Gary Kildal's PLM. I have a version of
> XPL that translated XPL to C to build apps, and a MODCOM III simulator
> written in XPL. Runs on Windows and Linux. I needed a reference for the
> MODCOMP III to use as a guide to a possible emulation of one that had been
> used as a front-end processor for the CDC 6500 at Purdue. Their unique
> home-rolled OS came to us as source code, and with some effort, I got it to
> build and run. It was the first computer system I used, both punched-card
> and terminal input. The terminal experience wasn't like that of Control
> Data's offerings.
>
> So you're saying this XPL is written in XPL? Maybe use this XPL-to-C
> compiler to bootstrap it?
>
> -Dai
>
> On Mon, Sep 15, 2025 at 12:42 PM Eric Swenson <er...@sw...> wrote:
>
>> It’s going to take some work. Doesn’t compile. And I’m not sure how to
>> build the XPL compiler from source (xcom.xpl).
>>
>> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>>
>> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working
>> on” means that I gather all the right bits, make sure it compiles and runs,
>> updating it if necessary until it does.  Lots of the MIT-MULTICS code,
>> based on MR10.2, needs updating for releases post-MR10.2.
>>
>> I’ll reply to this thread regarding the results of working on XPL.
>>
>> — Eric
>>
>> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...>
>> wrote:
>>
>> Eric:
>>
>> I would be interested in the XPL related stuff. It was the first
>> "compiler generator" language I ran across and wanted to play with badly
>> but never had access to a system that ran it.
>>
>> It's a stripped down version of PL/1 with some tools for generating what
>> are really translators rather than compilers (though with the addition of
>> semantic translation code you can make a compiler with it).
>>
>> Not great urgency but I'd love to play with it.
>>
>> Thanks,
>> Dean
>>
>> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...> wrote:
>>
>>> Attached you’ll find a list of all the programs installed in the >aml
>>> directory on MIT-MULTICS at the time that that system was shut down.  Those
>>> entries marked with INSTALLED have been installed in the >aml for MR12.9
>>> (not yet released).  Those entries not so-marked have not been included.
>>>
>>> I’m writing to provide anyone with the opportunity to lobby for other
>>> programs to get integrated into the MR12.9 AML library.  Please let me know
>>> if you see something here that you really want installed in MR12.9 >aml.
>>> Feel free to ask me questions about these and I’ll endeavor to answer them
>>> (undoubtedly it will require me to look at the sources/info segments and
>>> attempt to rationalize not including them or including them).
>>>
>>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of
>>> which I’ve already installed a few programs into the MR12.9 >aml.
>>>
>>> — Eric
>>>
>>> _______________________________________________
>>> Dps8m-developers mailing list
>>> Dps...@li...
>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>>
>>
>>
>> _______________________________________________
>> Dps8m-developers mailing list
>> Dps...@li...
>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>
> _______________________________________________
> Dps8m-users mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-users
>

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Eric S. <er...@sw...>]

Actually, this version of XPL is written in PL/1 and assembler, but the compiler (xcom) is written in XPL. So it would appear you need xcom to compile itself from source (xcom.xpl). I tried compiling the compiler using MIT’s xcom executable, and got oodles of errors.

> On Sep 15, 2025, at 18:30, Daiyu Hurst <dai...@gm...> wrote:
> 
> So you're saying this XPL is written in XPL? Maybe use this XPL-to-C compiler to bootstrap it?

Re: [dps8m-developers] MIT-MULTICS SIPB Library and MR12.9

[Eric S. <er...@sw...>]

<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div dir="ltr"></div><div dir="ltr">Packaging the archive would mean making a dump tape of the stuff from MIT-MULTICS and making that available. I have no issue doing that. However, most users will not want the pain of having to load the tape somewhere, deal with the MR10.2-MR12.9 differences, rebuild everything and then have to deal with installing it somewhere. Most Multics users these days will find this a big headache. I’m happy to give such a tape to you and others who want it.</div><div dir="ltr"><br></div><div dir="ltr">However, in the throes of releasing MR12.9 and populating &gt;aml with working stuff, correctly installed and with appropriate documentation. Most users will appreciate that form of delivery. It will mean ready-to-use programs in standard system libraries that will be in QuickStart systems.</div><div dir="ltr"><br></div><div dir="ltr">So at this point I’m soliciting recommendations for inclusion in &gt;aml for MR12.9.</div><div dir="ltr"><br></div><div dir="ltr">Let me know if you want a copy of the raw stuff, Daiyu, and I can make a dump tape of it.</div><div dir="ltr"><br><blockquote type="cite">On Sep 15, 2025, at 18:31, Daiyu Hurst &lt;dai...@gm...&gt; wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-size:large">Some of these look interesting. Might it not be easier to just package that archive in a way that a Multics administrator could handle their&nbsp;own 10.2-to-16.9 ports?</div><div class="gmail_default" style="font-size:large"><br></div><div class="gmail_default" style="font-size:large">-Dai</div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Mon, Sep 15, 2025 at 12:31 PM Eric Swenson &lt;<a href="mailto:er...@sw...">er...@sw...</a>&gt; wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Attached you’ll find a list of all the programs installed in the SIPB library on MIT-MULTICS at the time that that system was shut down.&nbsp; Those (few) entries marked as INSTALLED have been installed in the &gt;aml library for MR12.9 (not yet released).<br>
<br>
If anyone is particularly interested in having any of these installed in MR12.9 &gt;aml, please let me know.&nbsp; Attached is also a listing of the CURRENT contents of &gt;aml on GHM — the system from which the MR12.9 release will be created — for your reference.<br>
<br>
— Eric<br>
<br>
_______________________________________________<br>
Dps8m-developers mailing list<br>
<a href="mailto:Dps...@li..." target="_blank">Dps...@li...</a><br>
<a href="https://lists.sourceforge.net/lists/listinfo/dps8m-developers" rel="noreferrer" target="_blank">https://lists.sourceforge.net/lists/listinfo/dps8m-developers</a><br>
</blockquote></div>
</div></blockquote></body></html>

Re: [dps8m-developers] MIT-MULTICS SIPB Library and MR12.9

[Daiyu H. <dai...@gm...>]

Some of these look interesting. Might it not be easier to just package that
archive in a way that a Multics administrator could handle their own
10.2-to-16.9 ports?

-Dai

On Mon, Sep 15, 2025 at 12:31 PM Eric Swenson <er...@sw...> wrote:

> Attached you’ll find a list of all the programs installed in the SIPB
> library on MIT-MULTICS at the time that that system was shut down.  Those
> (few) entries marked as INSTALLED have been installed in the >aml library
> for MR12.9 (not yet released).
>
> If anyone is particularly interested in having any of these installed in
> MR12.9 >aml, please let me know.  Attached is also a listing of the CURRENT
> contents of >aml on GHM — the system from which the MR12.9 release will be
> created — for your reference.
>
> — Eric
>
> _______________________________________________
> Dps8m-developers mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Daiyu H. <dai...@gm...>]

XPL is kinda cool, it reminds me of Gary Kildal's PLM. I have a version of
XPL that translated XPL to C to build apps, and a MODCOM III simulator
written in XPL. Runs on Windows and Linux. I needed a reference for the
MODCOMP III to use as a guide to a possible emulation of one that had been
used as a front-end processor for the CDC 6500 at Purdue. Their unique
home-rolled OS came to us as source code, and with some effort, I got it to
build and run. It was the first computer system I used, both punched-card
and terminal input. The terminal experience wasn't like that of Control
Data's offerings.

So you're saying this XPL is written in XPL? Maybe use this XPL-to-C
compiler to bootstrap it?

-Dai

On Mon, Sep 15, 2025 at 12:42 PM Eric Swenson <er...@sw...> wrote:

> It’s going to take some work. Doesn’t compile. And I’m not sure how to
> build the XPL compiler from source (xcom.xpl).
>
> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>
> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on”
> means that I gather all the right bits, make sure it compiles and runs,
> updating it if necessary until it does.  Lots of the MIT-MULTICS code,
> based on MR10.2, needs updating for releases post-MR10.2.
>
> I’ll reply to this thread regarding the results of working on XPL.
>
> — Eric
>
> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...> wrote:
>
> Eric:
>
> I would be interested in the XPL related stuff. It was the first "compiler
> generator" language I ran across and wanted to play with badly but never
> had access to a system that ran it.
>
> It's a stripped down version of PL/1 with some tools for generating what
> are really translators rather than compilers (though with the addition of
> semantic translation code you can make a compiler with it).
>
> Not great urgency but I'd love to play with it.
>
> Thanks,
> Dean
>
> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...> wrote:
>
>> Attached you’ll find a list of all the programs installed in the >aml
>> directory on MIT-MULTICS at the time that that system was shut down.  Those
>> entries marked with INSTALLED have been installed in the >aml for MR12.9
>> (not yet released).  Those entries not so-marked have not been included.
>>
>> I’m writing to provide anyone with the opportunity to lobby for other
>> programs to get integrated into the MR12.9 AML library.  Please let me know
>> if you see something here that you really want installed in MR12.9 >aml.
>> Feel free to ask me questions about these and I’ll endeavor to answer them
>> (undoubtedly it will require me to look at the sources/info segments and
>> attempt to rationalize not including them or including them).
>>
>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of
>> which I’ve already installed a few programs into the MR12.9 >aml.
>>
>> — Eric
>>
>> _______________________________________________
>> Dps8m-developers mailing list
>> Dps...@li...
>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>
>
>
> _______________________________________________
> Dps8m-developers mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>

Re: [dps8m-developers] XPL

[Eric S. <er...@sw...>]

<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div dir="ltr"></div><div dir="ltr">Hi Dean. I created an “xpl” directory under your home directory on GHM with the XPL sources and objects. There is also the “xcom” compiler executable from MIT-MULTICS since I’ve been unable to build it so far. It looks like there are a lot of H645 dependencies, but I may be wrong.</div><div dir="ltr"><br></div><div dir="ltr">The sources are slightly modified from MIT-MULTICS versions — so that they would compile/assemble.</div><div dir="ltr"><br></div><div dir="ltr">Let me know if I can help in any way.</div><div dir="ltr"><br></div><div dir="ltr">— Eric</div><div dir="ltr"><br><blockquote type="cite">On Sep 15, 2025, at 11:12, Dean Anderson &lt;de...@go...&gt; wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr"><div dir="ltr">Eric:<div><br></div><div>Yes, please make it available so I can at least look at it.</div><div><br></div><div>Thanks,</div><div>Dean</div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Mon, Sep 15, 2025 at 12:50 PM Eric Swenson &lt;<a href="mailto:er...@sw...">er...@sw...</a>&gt; wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>While I’ve made progress, and can compile XPL, I’m unable to compile the compiler (xcom) with the generated xpl executable.&nbsp; I tried compiling with the xpl executable in &gt;aml from MIT-MULTICS, and that gets the same errors as my recompiled version.<div><br></div><div>There are many references to H645 in the code, that I suspect it doesn’t work (never worked) on the H6180 (or DPS8/M).&nbsp; I’m going to give up for now, although if you’d like to take a look, Dean, I can make all this available to you on GHM.</div><div><br></div><div>— Eric<br id="m_-3514638666509088356lineBreakAtBeginningOfMessage"><div><br><blockquote type="cite"><div>On Sep 15, 2025, at 9:41 AM, Eric Swenson &lt;<a href="mailto:er...@sw..." target="_blank">er...@sw...</a>&gt; wrote:</div><br><div><div>It’s going to take some work. Doesn’t compile. And I’m not sure how to build the XPL compiler from source (xcom.xpl).<br id="m_-3514638666509088356lineBreakAtBeginningOfMessage"><div><br><blockquote type="cite"><div>On Sep 15, 2025, at 9:32 AM, Eric Swenson &lt;<a href="mailto:er...@sw..." target="_blank">er...@sw...</a>&gt; wrote:</div><br><div><div>Ok.&nbsp; Thanks for the suggestion, Dean. I’ll work on XPL next. &nbsp;“Working on” means that I gather all the right bits, make sure it compiles and runs, updating it if necessary until it does.&nbsp; Lots of the MIT-MULTICS code, based on MR10.2, needs updating for releases post-MR10.2.<div><br></div><div>I’ll reply to this thread regarding the results of working on XPL.</div><div><br></div><div>— Eric<br id="m_-3514638666509088356lineBreakAtBeginningOfMessage"><div><br><blockquote type="cite"><div>On Sep 15, 2025, at 9:25 AM, Dean Anderson &lt;<a href="mailto:de...@go..." target="_blank">de...@go...</a>&gt; wrote:</div><br><div><div dir="ltr">Eric:<div><br></div><div>I would be interested in the XPL related stuff. It was the first "compiler generator" language I ran across and wanted to play with badly but never had access to a system that ran it.</div><div><br></div><div>It's a stripped down version of PL/1 with some tools for generating what are really translators rather than compilers (though with the addition&nbsp;of semantic translation code you can make a compiler with it).</div><div><br></div><div>Not great urgency but I'd love to play with it.</div><div><br></div><div>Thanks,</div><div>Dean</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson &lt;<a href="mailto:er...@sw..." target="_blank">er...@sw...</a>&gt; wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Attached you’ll find a list of all the programs installed in the &gt;aml directory on MIT-MULTICS at the time that that system was shut down.&nbsp; Those entries marked with INSTALLED have been installed in the &gt;aml for MR12.9 (not yet released).&nbsp; Those entries not so-marked have not been included.<br>
<br>
I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.&nbsp; Please let me know if you see something here that you really want installed in MR12.9 &gt;aml.&nbsp; Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).<br>
<br>
I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 &gt;aml.<br>
<br>
— Eric<br>
<br>
_______________________________________________<br>
Dps8m-developers mailing list<br>
<a href="mailto:Dps...@li..." target="_blank">Dps...@li...</a><br>
<a href="https://lists.sourceforge.net/lists/listinfo/dps8m-developers" rel="noreferrer" target="_blank">https://lists.sourceforge.net/lists/listinfo/dps8m-developers</a><br>
</blockquote></div>
</div></blockquote></div><br></div></div></div></blockquote></div><br></div></div></blockquote></div><br></div></div></blockquote></div>
</div></blockquote></body></html>

Re: [dps8m-developers] XPL

[Dean A. <de...@go...>]

Eric:

Yes, please make it available so I can at least look at it.

Thanks,
Dean

On Mon, Sep 15, 2025 at 12:50 PM Eric Swenson <er...@sw...> wrote:

> While I’ve made progress, and can compile XPL, I’m unable to compile the
> compiler (xcom) with the generated xpl executable.  I tried compiling with
> the xpl executable in >aml from MIT-MULTICS, and that gets the same errors
> as my recompiled version.
>
> There are many references to H645 in the code, that I suspect it doesn’t
> work (never worked) on the H6180 (or DPS8/M).  I’m going to give up for
> now, although if you’d like to take a look, Dean, I can make all this
> available to you on GHM.
>
> — Eric
>
> On Sep 15, 2025, at 9:41 AM, Eric Swenson <er...@sw...> wrote:
>
> It’s going to take some work. Doesn’t compile. And I’m not sure how to
> build the XPL compiler from source (xcom.xpl).
>
> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>
> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on”
> means that I gather all the right bits, make sure it compiles and runs,
> updating it if necessary until it does.  Lots of the MIT-MULTICS code,
> based on MR10.2, needs updating for releases post-MR10.2.
>
> I’ll reply to this thread regarding the results of working on XPL.
>
> — Eric
>
> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...> wrote:
>
> Eric:
>
> I would be interested in the XPL related stuff. It was the first "compiler
> generator" language I ran across and wanted to play with badly but never
> had access to a system that ran it.
>
> It's a stripped down version of PL/1 with some tools for generating what
> are really translators rather than compilers (though with the addition of
> semantic translation code you can make a compiler with it).
>
> Not great urgency but I'd love to play with it.
>
> Thanks,
> Dean
>
> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...> wrote:
>
>> Attached you’ll find a list of all the programs installed in the >aml
>> directory on MIT-MULTICS at the time that that system was shut down.  Those
>> entries marked with INSTALLED have been installed in the >aml for MR12.9
>> (not yet released).  Those entries not so-marked have not been included.
>>
>> I’m writing to provide anyone with the opportunity to lobby for other
>> programs to get integrated into the MR12.9 AML library.  Please let me know
>> if you see something here that you really want installed in MR12.9 >aml.
>> Feel free to ask me questions about these and I’ll endeavor to answer them
>> (undoubtedly it will require me to look at the sources/info segments and
>> attempt to rationalize not including them or including them).
>>
>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of
>> which I’ve already installed a few programs into the MR12.9 >aml.
>>
>> — Eric
>>
>> _______________________________________________
>> Dps8m-developers mailing list
>> Dps...@li...
>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>>
>
>
>
>

Re: [dps8m-developers] XPL

[Eric S. <er...@sw...>]

While I’ve made progress, and can compile XPL, I’m unable to compile the compiler (xcom) with the generated xpl executable.  I tried compiling with the xpl executable in >aml from MIT-MULTICS, and that gets the same errors as my recompiled version.

There are many references to H645 in the code, that I suspect it doesn’t work (never worked) on the H6180 (or DPS8/M).  I’m going to give up for now, although if you’d like to take a look, Dean, I can make all this available to you on GHM.

— Eric

> On Sep 15, 2025, at 9:41 AM, Eric Swenson <er...@sw...> wrote:
> 
> It’s going to take some work. Doesn’t compile. And I’m not sure how to build the XPL compiler from source (xcom.xpl).
> 
>> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
>> 
>> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on” means that I gather all the right bits, make sure it compiles and runs, updating it if necessary until it does.  Lots of the MIT-MULTICS code, based on MR10.2, needs updating for releases post-MR10.2.
>> 
>> I’ll reply to this thread regarding the results of working on XPL.
>> 
>> — Eric
>> 
>>> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...> wrote:
>>> 
>>> Eric:
>>> 
>>> I would be interested in the XPL related stuff. It was the first "compiler generator" language I ran across and wanted to play with badly but never had access to a system that ran it.
>>> 
>>> It's a stripped down version of PL/1 with some tools for generating what are really translators rather than compilers (though with the addition of semantic translation code you can make a compiler with it).
>>> 
>>> Not great urgency but I'd love to play with it.
>>> 
>>> Thanks,
>>> Dean
>>> 
>>> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>>>> Attached you’ll find a list of all the programs installed in the >aml directory on MIT-MULTICS at the time that that system was shut down.  Those entries marked with INSTALLED have been installed in the >aml for MR12.9 (not yet released).  Those entries not so-marked have not been included.
>>>> 
>>>> I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.  Please let me know if you see something here that you really want installed in MR12.9 >aml.  Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).
>>>> 
>>>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 >aml.
>>>> 
>>>> — Eric
>>>> 
>>>> _______________________________________________
>>>> Dps8m-developers mailing list
>>>> Dps...@li... <mailto:Dps...@li...>
>>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>> 
> 

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Eric S. <er...@sw...>]

It’s going to take some work. Doesn’t compile. And I’m not sure how to build the XPL compiler from source (xcom.xpl).

> On Sep 15, 2025, at 9:32 AM, Eric Swenson <er...@sw...> wrote:
> 
> Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on” means that I gather all the right bits, make sure it compiles and runs, updating it if necessary until it does.  Lots of the MIT-MULTICS code, based on MR10.2, needs updating for releases post-MR10.2.
> 
> I’ll reply to this thread regarding the results of working on XPL.
> 
> — Eric
> 
>> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...> wrote:
>> 
>> Eric:
>> 
>> I would be interested in the XPL related stuff. It was the first "compiler generator" language I ran across and wanted to play with badly but never had access to a system that ran it.
>> 
>> It's a stripped down version of PL/1 with some tools for generating what are really translators rather than compilers (though with the addition of semantic translation code you can make a compiler with it).
>> 
>> Not great urgency but I'd love to play with it.
>> 
>> Thanks,
>> Dean
>> 
>> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>>> Attached you’ll find a list of all the programs installed in the >aml directory on MIT-MULTICS at the time that that system was shut down.  Those entries marked with INSTALLED have been installed in the >aml for MR12.9 (not yet released).  Those entries not so-marked have not been included.
>>> 
>>> I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.  Please let me know if you see something here that you really want installed in MR12.9 >aml.  Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).
>>> 
>>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 >aml.
>>> 
>>> — Eric
>>> 
>>> _______________________________________________
>>> Dps8m-developers mailing list
>>> Dps...@li... <mailto:Dps...@li...>
>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
> 

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Eric S. <er...@sw...>]

Ok.  Thanks for the suggestion, Dean. I’ll work on XPL next.  “Working on” means that I gather all the right bits, make sure it compiles and runs, updating it if necessary until it does.  Lots of the MIT-MULTICS code, based on MR10.2, needs updating for releases post-MR10.2.

I’ll reply to this thread regarding the results of working on XPL.

— Eric

> On Sep 15, 2025, at 9:25 AM, Dean Anderson <de...@go...> wrote:
> 
> Eric:
> 
> I would be interested in the XPL related stuff. It was the first "compiler generator" language I ran across and wanted to play with badly but never had access to a system that ran it.
> 
> It's a stripped down version of PL/1 with some tools for generating what are really translators rather than compilers (though with the addition of semantic translation code you can make a compiler with it).
> 
> Not great urgency but I'd love to play with it.
> 
> Thanks,
> Dean
> 
> On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw... <mailto:er...@sw...>> wrote:
>> Attached you’ll find a list of all the programs installed in the >aml directory on MIT-MULTICS at the time that that system was shut down.  Those entries marked with INSTALLED have been installed in the >aml for MR12.9 (not yet released).  Those entries not so-marked have not been included.
>> 
>> I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.  Please let me know if you see something here that you really want installed in MR12.9 >aml.  Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).
>> 
>> I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 >aml.
>> 
>> — Eric
>> 
>> _______________________________________________
>> Dps8m-developers mailing list
>> Dps...@li... <mailto:Dps...@li...>
>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers

[dps8m-developers] MIT-MULTICS SIPB Library and MR12.9

[Eric S. <er...@sw...>]

Attached you’ll find a list of all the programs installed in the SIPB library on MIT-MULTICS at the time that that system was shut down.  Those (few) entries marked as INSTALLED have been installed in the >aml library for MR12.9 (not yet released).

If anyone is particularly interested in having any of these installed in MR12.9 >aml, please let me know.  Attached is also a listing of the CURRENT contents of >aml on GHM — the system from which the MR12.9 release will be created — for your reference.

— Eric

Re: [dps8m-developers] MIT-MULTICS AML and MR12.9

[Dean A. <de...@go...>]

Eric:

I would be interested in the XPL related stuff. It was the first "compiler
generator" language I ran across and wanted to play with badly but never
had access to a system that ran it.

It's a stripped down version of PL/1 with some tools for generating what
are really translators rather than compilers (though with the addition of
semantic translation code you can make a compiler with it).

Not great urgency but I'd love to play with it.

Thanks,
Dean

On Mon, Sep 15, 2025 at 10:45 AM Eric Swenson <er...@sw...> wrote:

> Attached you’ll find a list of all the programs installed in the >aml
> directory on MIT-MULTICS at the time that that system was shut down.  Those
> entries marked with INSTALLED have been installed in the >aml for MR12.9
> (not yet released).  Those entries not so-marked have not been included.
>
> I’m writing to provide anyone with the opportunity to lobby for other
> programs to get integrated into the MR12.9 AML library.  Please let me know
> if you see something here that you really want installed in MR12.9 >aml.
> Feel free to ask me questions about these and I’ll endeavor to answer them
> (undoubtedly it will require me to look at the sources/info segments and
> attempt to rationalize not including them or including them).
>
> I will soon send out a similar list for the MIT-MULTICS SIPB library, of
> which I’ve already installed a few programs into the MR12.9 >aml.
>
> — Eric
>
> _______________________________________________
> Dps8m-developers mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>

[dps8m-developers] MIT-MULTICS AML and MR12.9

[Eric S. <er...@sw...>]

Attached you’ll find a list of all the programs installed in the >aml directory on MIT-MULTICS at the time that that system was shut down.  Those entries marked with INSTALLED have been installed in the >aml for MR12.9 (not yet released).  Those entries not so-marked have not been included.

I’m writing to provide anyone with the opportunity to lobby for other programs to get integrated into the MR12.9 AML library.  Please let me know if you see something here that you really want installed in MR12.9 >aml.  Feel free to ask me questions about these and I’ll endeavor to answer them (undoubtedly it will require me to look at the sources/info segments and attempt to rationalize not including them or including them).

I will soon send out a similar list for the MIT-MULTICS SIPB library, of which I’ve already installed a few programs into the MR12.9 >aml.

— Eric

Re: [dps8m-developers] [Dps8m-users] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[Jeff J. <tr...@po...>]

I approve this MCR

--
Jeffrey H. Johnson
tr...@po...

On Mon, Sep 8, 2025, at 6:52 PM, Eric Swenson wrote:
> Hello Multicians:
> 
> Ticket https://multics-trac.swenson.org/ticket/381#ticket describes an issue with the “attach” command of the GCOS daemon.  This command doesn’t allow specification of the card reader to attach, and simply uses the “rdra” card reader.  Since this card reader may be in use by the IO SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon card reader driver is logged out.
> 
> This MCR proposes to address this problem by adding a parameter to the “attach” command.
> 
> Please review MCR10159 and comment on it or indicate your approval for this set of changes.  Provide comments or approvals by Reply-All to this email.
>  
> Thanks in advance for your review.
>  
> REVIEW DOCUMENT:         MCR10159: https://s3.amazonaws.com/eswenson-multics/public/mcrs/MCR10159.pdf
> 
> REVIEW ENDS:             Monday, August 22, 2025
> 
> _______________________________________________
> Dps8m-users mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-users
> 

Re: [dps8m-developers] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[Dean A. <de...@go...>]

I very much approve this MCR.

Dean S. Anderson

On Mon, Sep 8, 2025 at 5:53 PM Eric Swenson <er...@sw...> wrote:

> Hello Multicians:
>
>
> Ticket https://multics-trac.swenson.org/ticket/381#ticket describes an
> issue with the “attach” command of the GCOS daemon.  This command doesn’t
> allow specification of the card reader to attach, and simply uses the
> “rdra” card reader.  Since this card reader may be in use by the IO
> SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon
> card reader driver is logged out.
>
>
> This MCR proposes to address this problem by adding a parameter to the
> “attach” command.
>
>
> Please review MCR10159 and comment on it or indicate your approval for
> this set of changes.  Provide comments or approvals by Reply-All to this
> email.
>
>
>
> Thanks in advance for your review.
>
>
>
> REVIEW DOCUMENT:          MCR10159:
> https://s3.amazonaws.com/eswenson-multics/public/mcrs/MCR10159.pdf
>
>
> REVIEW ENDS:              Monday, August 22, 2025
> _______________________________________________
> Dps8m-developers mailing list
> Dps...@li...
> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>

Re: [dps8m-developers] MCR10159: GCOS Daemon should support card reader device name in "attach" command

[Eric S. <er...@sw...>]

It occurred to me that you might be asking what happens with the proposed new code (rather than with the existing MR12.8 code).

The new code takes the tail of the command line, stripped of white space and tries to attach to the specified reader device. So “attach foo” will try to attach to a reader named “foo”. If no such reader is defined, an appropriate error is returned by the underlying IO attachment system.  The proposed code doesn’t parse the tail of the command. It just strips white space off both ends and passes to the IO attachment API.

> On Sep 8, 2025, at 20:33, Eric Swenson <er...@sw...> wrote:
> 
> The gcos daemon software doesn’t parse anything past the command name.
> 
>> On Sep 8, 2025, at 20:15, Peter Flass <Pe...@ir...> wrote:
>> 
>> What do you get if you specify an invalid argument?
>> 
>>>> On 9/8/25 15:52, Eric Swenson wrote:
>>> Hello Multicians:
>>> Ticket https://multics-trac.swenson.org/ticket/381#ticket <https:// multics-trac.swenson.org/ticket/381#ticket> describes an issue with the “attach” command of the GCOS daemon.  This command doesn’t allow specification of the card reader to attach, and simply uses the “rdra” card reader.  Since this card reader may be in use by the IO SysDaemon, it is not possible to use the GCOS daemon until the IO SysDaemon card reader driver is logged out.
>>> This MCR proposes to address this problem by adding a parameter to the “attach” command.
>>> Please review MCR10159 and comment on it or indicate your approval for this set of changes.  Provide comments or approvals by Reply-All to this email.
>>> Thanks in advance for your review.
>>> REVIEW DOCUMENT: MCR10159: https://s3.amazonaws.com/eswenson-multics/ public/mcrs/MCR10159.pdf <https://s3.amazonaws.com/eswenson-multics/ public/mcrs/MCR10159.pdf>
>>> REVIEW ENDS: Monday, August 22, 2025
>>> _______________________________________________
>>> Dps8m-developers mailing list
>>> Dps...@li...
>>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers
>> 
>> 
>> 
>> _______________________________________________
>> Dps8m-developers mailing list
>> Dps...@li...
>> https://lists.sourceforge.net/lists/listinfo/dps8m-developers